Our Town Evaluation Model

During our trip, we will be applying a tool we are developing called 'Decision Apprentice'. Currently, this tool is available as software that can be installed and used on any Windows PC. We are developing a Silverlight version that will be available as a web service, but this will not be released until Silverlight 3 is available, probably this fall.

In the meantime, I will assess each town we visit using the version of the tool installed on my laptop. I will post the results of each assessment on the web, with links here.

First, a little information on the decision model we've built. I will cut and paste the basic template here, though it is much more difficult to see what is going on in this html version -- it is easier to understand the different levels of the tree when you can expand and collapse each section as you go. Unfortunately, I haven't figured out yet how to get this functionality into my blog entry...

There are five high level branches of the decision: climate, cost of living, work opportunities, quality of life and accessibility. Here is the total model, and I will explain the five branches further after the tree:

• ○This town is a good choice for our next home.




• ○AND [1 of 5] : This town has a good climate.




• ○AND [1 of 2] : This town has moderate winters.




• ○OR [1 of 2] : The number of days each year at or below freezing should be 10 or less.


• ○OR [2 of 2] : The average winter temperature should be at least 10 degrees warmer than Hudson OH.


• ○AND [2 of 2] : This town has a good balance of sunshine and summer heat.




• ○OR [1 of 2] : This town has moderate summers in exchange for lots of sunshine.


• ○OR [2 of 2] : This town has lots of sunshine but hot summers.


• ○AND [2 of 5] : This town has an affordable cost of living.




• ○AND [1 of 4] : This town has health insurance available for less than $1500 a month.


• ○AND [2 of 4] : This town has lower taxes than my current home town.


• ○AND [3 of 4] : This town has a lower cost of living index than my current home town.


• ○AND [4 of 4] : This town has good housing for less than my maximum house budget.


• ○AND [3 of 5] : This town has good work opportunities.




• ○OR [1 of 2] : This town has good fulltime work opportunities.


• ○OR [2 of 2] : This town has good freelance work opportunities.


• ○AND [4 of 5] : This town provides a good quality of life.




• ○AND [1 of 2] : This town can provide a healthy lifestyle.




• ○OR [1 of 2] : This town has neighborhoods that provide a walkable urban lifestyle.




• ○AND [1 of 2] : This town has at least one neighborhood with a walkscore greater than 90.


• ○AND [2 of 2] : The neighborhood with a walkscore greater than 90 has a crime rate no greater than 125% of the national average.


• ○OR [2 of 2] : This town has areas with affordable nature 'compounds'.


• ○AND [2 of 2] : This town can provide an inspirational lifestyle.




• ○OR [1 of 3] : This town has a high quality university.


• ○OR [2 of 3] : This town has cultural diversity.


• ○OR [3 of 3] : This town has beautiful scenery and views.


• ○AND [5 of 5] : This town has good accessibility.




• ○AND [1 of 2] : This town is within 2 hours of a major airport.


• ○AND [2 of 2] : This town is within 30 minutes of high quality hospitals.

When this tree is filled out for an actual town, certain parts of the text will be replaced with specifics (such as 'Memphis'), and the green and red icons will indicate if the branch is true or false. The actual application looks like this:

I collect information on the internet, and copy notes, pdfs and pictures into the appropriate item in the tree. I set the values of the lowest level items to true or false, based on what I learn, but the application 'reasons up' the tree, setting higher level items to true or false based on rules of logic.

In my model, all five high level decisions must be true for the top item - this town is a good choice for our new home - to be true. Our own preferences are built into each of the five branches, to establish what is a 'good' climate, quality of life, etc. It was quick and easy to build this tree, and now I can apply the model consistently to each place we visit. I have bookmarked a number of sites that have the information I need to evaluate each item in the tree.

So far I have evaluated Lexington and Memphis, our first two stops, and sad to say - neither one of them pass the test. I will post these two applications of the model in my next blog entries, and talk about why each fails. Will any place we visit pass? Or is the decision model too tough? Will we need to compromise, and if so, on which items? That is part of what our walkabout is meant to show us.